Computer Vision Systems
8th International Conference, ICVS 2011, Sophia Antipolis, France, September 20-22, 2011, Proceedings
Published on 22. September 2011
XI, 223 pages
978-3-642-23968-7 (ISBN)
Description
This book constitutes the refereed proceedings of the 8th International Conference on Computer Vision Systems, ICVS 2011, held in Sophia Antipolis, France, in September 2009. The 22 revised papers presented were carefully reviewed and selected from 58 submissions. The papers are organized in topical sections on vision systems, control of perception, performance evaluation, activity recognition, and knowledge directed vision.
More details
Other editions
Additional editions
James L. Crowley | Bruce Draper | Monique Thonnat
Computer Vision Systems
8th International Conference, ICVS 2011, Sophia Antipolis, France, September 20-22, 2011, Proceedings
Book
09/2011
1st Edition
Springer
€53.49
Shipment within 7-9 days
Content
- Title
- Preface
- Organization
- Table of Contents
- Vision System
- Knowing What Happened - Automatic Documentation of Image Analysis Processes
- Introduction
- Related Work
- Automatic Documentation
- Operators
- Processing Graph
- Implementation
- External Representation of the Processing History
- Exploring the Processing History: Chipory
- Alida in Practice: The MiToBo Image Processing Toolbox
- Conclusion
- References
- Efficient Use of Geometric Constraints for Sliding-Window Object Detection in Video
- Introduction
- The Space of Valid Object Detections
- Detection Algorithms
- Experimental Results
- Conclusion
- References
- A Method for Asteroids 3D Surface Reconstruction from Close Approach Distances
- Introduction
- Feature Detection
- Surface Reconstruction
- Affine Camera Approximation
- Affine Structure from Motion
- Optimisation with Augmented Lagrange Multipliers (ALM)
- Results
- Conclusions and Future Work
- References
- RT-SLAM: A Generic and Real-Time Visual SLAM Implementation
- Motivation
- Overall Architecture
- Inertial/Visual SLAM within RT-SLAM
- Active Search and One-Point RANSAC
- Landmark Parametrizations and Reparametrization
- Motion Prediction
- Image Processing
- Results
- Constant Velocity Model
- Inertial/Visual SLAM
- Outlook
- References
- Control of Perception (I)
- A Quantitative Comparison of Speed and Reliability for Log-Polar Mapping Techniques
- Introduction
- Log-Polar Mapping: The Blind-Spot Model
- Comparison and Results
- Computational Load
- Image Quality Indexes
- Conclusions and Future Work
- References
- Toward Accurate Feature Detectors Performance Evaluation
- Introduction
- Repeatability Score Analysis
- Definitions
- Issue 1: Implicit Assumption About Biunique Point-to-Point Correspondence
- Issue 2: Noisy Inputs Produce High Scores
- Issue 3: Repeatability Score Depends on Scale Factor
- Alternative Repeatability Score Definitions
- Remaining Issues of Repeatability Score Definition
- Proposed Modifications toward More Accurate Repeatability Score Calculation
- Complex Shape of the -Neighborhood Region
- New ``Correspondence'' Assumptions and Two Types of Repeatability Score
- Conclusion
- References
- Evaluation of Local Descriptors for Action Recognition in Videos
- Introduction
- Evaluation Framework
- Space-Time Local Features
- Bag-of-Words Model
- Classification
- Evaluation
- Experiments
- Weizmann Action Recognition Dataset
- KTH Dataset
- ADL Dataset
- Keck Dataset
- Conclusions
- References
- Performance Evaluation (II)
- On the Spatial Extents of SIFT Descriptors for Visual Concept Detection
- Introduction
- Related Work
- Concept Detection System
- SIFT Descriptor Geometry
- Soft-Weighting Scheme
- Color and Spatial Information
- Kernel Choice
- Multiple Kernel Learning
- Experimental Results
- Evaluation Criteria
- Results
- Discussion
- Conclusions
- References
- An Experimental Framework for Evaluating PTZ Tracking Algorithms
- Introduction
- State of the Art
- Methodology
- PTZ Camera Calibration
- Tracking Algorithms for PTZ Camera
- Performance Evaluation
- Experiments
- Conclusions
- References
- Activity Recognition
- Unsupervised Activity Extraction on Long-Term Video Recordings Employing Soft Computing Relations
- Introduction
- Related Work
- General Overview of the System
- Real-Time Processing Object Detection and Tracking
- Detection
- Image Plane Tracking
- Multi-camera Fusion and 3D Tracking
- Data Preprocessing
- Activity Clustering Methodology
- Preliminary Definitions
- Clustering of Video Data
- Results and Evaluation
- Conclusions
- References
- Unsupervised Discovery, Modeling, and Analysis of Long Term Activities
- Introduction
- Related Work
- Actions
- Global Position and Speed
- Action Segments and Local Dynamics
- Action Descriptors
- Scene Context
- Learning a Topology
- Scene Model
- Activities
- Primitive Events
- Primitive Events Sequence
- Activity Discovery
- Activity Model
- Activity Recognition
- Experiments
- Evaluation
- Conclusions
- References
- Ontology-Based Realtime Activity Monitoring Using Beam Search
- Introduction
- Behaviour Modelling
- Initialising the Scene Interpretation System from the Ontology
- System Overview
- Rule Generation from the Ontology
- Interpretation Process
- Experimental Results and Evaluation
- Illustration of Probabilistic Rating
- Conclusions
- References
- Probabilistic Recognition of Complex Event
- Introduction
- Related Work
- Event Description Language
- Event Recognition Process
- Probabilistic Primitive State Recognition
- Hierarchical Uncertainty Propagation
- Experimental Results
- Conclusion
- References
- Control of Perception (I)
- Learning What Matters: Combining Probabilistic Models of 2D and 3D Saliency Cues
- Introduction
- Investigated Cues
- Surface Height Cue
- Relative Surface Orientation Cue
- Occluded Edges Cue
- Cue Combination
- Probabilistic Learning
- Evaluation and Results
- Conclusion and Future Work
- References
- 3D Saliency for Abnormal Motion Selection: The Role of the Depth Map
- Computational Attention
- Why 3D Features for Attention?
- Attention Model for Motion Selection
- Motion Features Extraction
- Spatio-Temporal Filtering of the Features
- From Feature Detection to Feature Selection
- Model Validation
- When Using 3D Features?
- Scenarios Used for Validation
- Validation of the Perspective Correction
- Scenarios Used for Validation
- Motion Speed Validation
- Discussion and Conclusion
- References
- Scene Understanding through Autonomous Interactive Perception
- Introduction
- Related Work
- System Overview
- Appearance Hypothesis
- Rigid Motion Hypothesis
- Motion Discrimination
- Experiments
- Conclusions
- References
- Knowledge Directed Vision
- A Cognitive Vision System for Nuclear Fusion Device Monitoring
- Introduction
- Proposed Approach
- Event Modeling for Vision Task Composing
- Application to Thermal Event Recognition at Tore Supra
- Hot Spot Detection
- Hot Spot Categorization
- Validation of Experimental Results
- Conclusion
- References
- Knowledge Representation and Inference for Grasp Affordances
- Introduction
- Overview
- Ontology of Concepts
- Knowledge Ontology Based on Textual Semantics
- Knowledge Ontology Based on Visual Features
- Knowledge Ontology Based on Conceptual Properties
- Knowledge Ontology Based on Grasp Affordances
- Knowledge Ontology Based on Part Functional Affordances
- Query Evaluation
- Detection of Part Affordances
- Detection of Grasp Affordances
- Query Matching
- Results and Evaluation
- Conclusion and Future Work
- References
- Control of Perception (II)
- Towards a General Abstraction Through Sequences of Conceptual Operations
- Introduction
- Previous Work
- Conceptual Operations
- Operations
- Match:
- Detect:
- Solve:
- Sequencing Operations
- Conclusion
- Towards a General Abstraction through Sequences of Conceptual Operations
- Introduction
- Previous Work
- Conceptual Operations
- Operations
- Sequencing Operations
- Conclusion
- References
- Girgit: A Dynamically Adaptive Vision System for Scene Understanding
- Introduction
- Related Work
- Foundations
- Computer Vision Systems
- Dynamically Adaptive Software Systems
- QoS Metrics
- The Girgit Dynamic Adaptation Framework
- Architecture
- Girgit's Implementation
- Example Execution in Girgit
- Empirical Evaluation
- Experimental Setup
- Results
- Discussion and Threats to Validity
- Conclusion
- References
- Run Time Adaptation of Video-Surveillance Systems: A Software Modeling Approach
- Introduction
- Software Models for Video-Surveillance Systems
- Using Software Models for Run Time Adaptation
- Software Architecture for Run Time Adaptation
- Example of a Run Time Adaptation Scenario
- Comparison with Other Works
- Conclusion
- References
- Automatically Searching for Optimal Parameter Settings Using a Genetic Algorithm
- Introduction
- Related Work
- Searching for Optimal Configurations
- Training, Tuning, and Test Datasets
- The LRPCA Baseline Algorithm
- Genetic Algorithm and Configuration Space
- The Optimal Configuration
- Data Mining the Search Space
- Conclusions
- References
- Author Index
